Persistent Identifier and Linking Infrastructure PILIN

1
Persistent Identifier and Linking Infrastructure
(PILIN)
PILIN is ARROW in Old Elvish

• within the ARROW2 project
• Nigel Ward, Technical Director
• 10 October 2006

2
DEST context for PILIN
3
PILIN objectives

• Pilot shared, standards-based, persistent
  identifier management infrastructure
• with the aim of
• Supporting adoption of
• persistent identifiers
• persistent identifier management services
• for Australian e-learning, e-research and
  e-science communities
• Planning for sustainable shared infrastructure
• persistence of identifiers
• persistence of identifier services
• over archival lengths of time

4
Australian experience with identifiers

• FRODO and MERRI repositories
• Many repository projects use persistent digital
  identifiers to manage large numbers of networked
  resources
• e-science, e-research, and e-learning
• Need for guidance and tools
• Emerging requirements
• Vocational Training, Schools, Transport Industry
• Wide range of identifiable resources
• web content, data sets, scientific instruments,
  services
• emerging uses eg semantic web abstractions,
  Parties
• National Library
• persistent identifiers information model that
  expresses relationships using identifiers
  interesting services

5
What does this experience tell us? (1)

• Identifiers are needed for
• more than resolving to a web location
• more than just web resources
• Need to go back to basics
• what are the characteristics of identifier?
• what do we mean by identifier, persistent
  identifier etc
• what are is the difference between an identifier
  and data associated with an identifier?

6
What does this experience tell us? (2)

• Need to decouple some information management
  concerns
• resource identification from resource location
• associative data about identifiers from metadata
  about the resource
• identifiers as abstractions from the technologies
  that instantiate them
• information models and architectures that take
  advantage of identifiers from the identifiers

7
What does this experience tell us? (3)

• Persistence is as much about
• governance
• persistence of the basic management and
  information services
• as it is about technology.
• Need to provide Guidance and Good Practice to
  projects
• Need good tools to help manage identifiers over
  their lifecycle
• Great potential to provide interesting
  information services over identifier enabled
  infrastructure
• enable identifier service mash-ups!?

8
What is identifier management infrastructure?

• Information infrastructure and services built
  using that infrastructure
• Core management identifier services
• Create, access, update
• Value-added resource management services
• Create version, Find appropriate copy, Move
  resource,
• The services
• are dynamic
• e.g. Find current location of resource
• require identifiers that are
• globally unique
• persistent over time
• should themselves provide reliable action over
  time

9
Why shared management infrastructure?

• Return on investment
• Co-develop tools and services
• Interoperate with other systemic information
  infrastructure
• Participating across a global networked
  environment
• e.g. global access to rapidly migrating and
  proliferating resources
• Assurance
• Shared responsibility for uniqueness of
  identifiers and persistence of identifiers and
  identifier services
• NB requires pro-active, trusted ownership and
  shared responsibility
• Scalability to millions (billions?) of
  identifiers

10
E-Framework approach

• Open standards based interfaces
• for interoperability
• In a service oriented environment
• The e-Framework analyses and documents SERVICE
  INTERFACES
• PILIN WILL DEVELOP SERVICE INTERFACE DEFINTIONS
  and TOOLKITS to support IDENTIFIER MANAGEMENT

11
Identifier Management Example note we know this
isnt right!!!
12
PILIN Summary
Standards
Timeline Sept 2006 - Dec 2007
E-Frameworks SUMs , Service Genres and
Expressions
Questions?
Policies
Ongoing Best Practice Support
Tools and Applications
Pilots Demonstrators
SUSTAINABLE SHARED INFRASTRUCTURE
Guidance and governance
This work is licensed under the Creative Commons
Attribution-NonCommercial ShareAlike 2.5 License.
To view a copy of this license, visit
http//creativecommons.org/licenses/by-nc-sa/2.5/a
u/ or send a letter to Creative Commons, 543
Howard Street, 5th Floor, San Francisco,
California, 94105, USA.
13
Why Handle based?

• Handles are used now by the MERRI and FRODO
  Projects
• Grid integration underway
• CNRI Handle System is an architecture, not just a
  registration and resolution service
• we can build other services on top of the core
  services
• Deals with non web resources
• Proven scalability
• Demonstrated ability to build a scalable Web
  Service layer on top
• Architecture allows Australia to be both part of
  global infrastructure and be independent
• Handle is one instantiation of the abstract
  identifier
• Want to integrate other required identifier
  services
• Get abstractions, practices and governance
  right - expect technology will migrate over
  time.